Intracavitary implants using radioactive sources is a keystone in the treatment of gynecological cancer with radiation. At present, dosimetric evaluation is inadequate for these implants using afterloading applicators such as the Fletcher-Suit ovoids, especially regarding dosages to the bladder and the rectum. Since these are the dose limiting organs, treatment decisions based on present evaluations are not optimal and could lead to increased incidences of normal tissue damage, and perhaps, of failures of tumor control (due to the reduction of prescribed tumor dose in the face of high organ doses). Further, reliable dose response analysis of treatment outcome and complications are impossible. The above inadequacies of intracavitary implant computerized dosimetry are due in part to two factors: 1) the failure to account for the attenuation effects of the tungsten shields, which are incorporated in the ovoids for the reduction of rectum and bladder doses; and 2) the inability of orthogonal x-ray projected images to adequately localize and delineate the rectum and the bladder in their entirety. This study proposes measurements of 3 dimensional dose distribution around Fletcher-Suit type applicators and incorporation of such data into computerized dose calculation. At present, 3D dose data are lacking, and, existing computer algorithms are incapable of accounting for the non-cylindrically symmetric dose distribution. Concomitantly, we shall explore the use of computerized tomography to determine the spatial relationship between the implanted sources and the bladder and the rectum, with techniques to minimize the "streaking" artifact caused by metal objects in the scanning field. In addition, a prospective study will evaluate the efficacy of the improved dosimetry in comparison with conventional computerized dose computations. The proposed method of dosimetric evaluation may improve the precision of tumor and organ doses. Ensuing treatment decisions based on such dose information may be more optimal and could lead to improved clinical results, i.e., reduced incidences of normal tissue damage and better local control probability. In addition, reliable doseresponse relationship of normal tissue tolerance may be determined in the future.